A neurocomputing approach to predict monsoon rainfall in monthly scale using SST anomaly as a predictor

A relationship between summer monsoon rainfall and sea surface temperature anomalies was investigated with the aim of predicting the monthly scale rainfall during the summer monsoon period over a section (80A degrees-90A degrees E, 14A degrees-24A degrees N) of eastern India that depends heavily upon the rainfall during the summer monsoon months for its agricultural practices. The association between area-averaged rainfall of June over the study zone and global sea surface temperature (SST) anomalies for the period 1982-2008 was examined and the variability of rainfall in monthly scale was calculated. With a view to significant variability in the rainfall in the monthly scale, it was decided to implement the artificial neural network (ANN) for forecasting the monthly scale rainfall using the SST anomalies as a predictor. Finally, the potential of ANN in this prediction has been assessed.